Portfolio

Data Science and Analytics Projects

Scraped 10-K Filings from the SEC website using BeautifulSoup and selected key sections for NLP Analysis to create a text based stock selection model based on an academic paper titled Lazy Prices.
Performed sentiment analysis on the 10-k's and evaluated the alpha factors by their Sharpe Ratio using the Alphalens library.

Implemented Using:
Python, BeautifulSoup, Alphalens, Pandas, Sklearn and Nltk Python libraries.

Stock Pairs Trading

Implemented Deep Q-Learning and defined the state, action and reward for a learning agent to be able to decide on trading actions for a pair of stocks.
Referred a research paper on forex trading pairs and used that as a baseline to code out the implementation strategy for a pair of stocks with the intention of maximizing returns.

Implemented Using:
Python, Keras, OpenAI, Numpy, Pandas, Reinforcement Learning

Fifa 19 Analysis

Various soccer related analyses using Naive Bayes, Logistic Regression, SVM and Clustering algorithms.
Developed an analysis tool to obtain like-for-like replacements for players based on various attributes and skill sets using clustering algorithms.

Implemented Using:
Python, Pandas, Numpy, Seaborn, Scikit-learn

Twitter Sentiment Anaysis

This app provides tweet sentiments for a user specified tweet topic and number of tweets the user would want to consider.

Implemented Using:
Python, Flask, Numpy, Tweepy, Google Cloud Platform

Flight Analysis using Tableau

Visualizations from the BTS data set that describe the on time performance of Spirit Airlines for the year 2018.

Java Backend Projects

QMoney

QMoney is a visual stock portfolio analyzer. It helps portfolio managers make trade recommendations for their clients.
- Implemented the core logic of the portfolio manager and published it as a library.
- Refactored code to add support for multiple stock quote services.
- Improved application stability and performance.

Implemented Using:
Java, REST API, Jackson, Interfaces, Code Refactoring, Gradle, Exception Handling, Multithreading.

QEats

QEats is a food ordering app that allows users to browse and order their favorite dishes from nearby restaurants. During the course of this Micro-Experience at Crio.do, developers build different parts of the QEats backend which is a Spring Boot application. Several REST API endpoints are implemented to query restaurant information and place food orders.

To give a sense of real-world problems, production issues are introduced for the developer to investigate using Scientific Debugging methods. Along with this, the developer is expected to improve the app performance under large load scenarios as well as include an advanced search feature in the app.

Implemented Using:
Spring Boot, Spring Data, REST API, Jackson, Mockito, JUnit, MongoDB, Scientific Debugging

Quiz App

A quiz application with an API endpoint that supports the following operations:
-GET /quiz/{moduleId}: fetches the quiz questions for a given module.
-POST /quiz/{moduleId}: submits user answers and returns an evaluation.

Implemented Using:
Spring Boot, REST API, MongoDB, JUnit